Add hdf5 inference preprocessing

Dockerfile

@@ -2,6 +2,12 @@ FROM sab148/misato-dataset:latest
 USER root
 # Set up time zone.
 
+RUN pip install gradio
+RUN pip install requests
+RUN pip install py3Dmol
+RUN pip install biopython
+RUN pip install pandas
+
 #RUN useradd -m -u 1000 user
 #USER user
 #ENV HOME=/home/user \
@@ -16,38 +22,10 @@ USER root
 #
 #CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
 
-# Set up a new user named "user" with user ID 1000
-
-
-RUN mkdir -p /maps
-WORKDIR /maps
-COPY maps/*pickle .
-
-
-RUN useradd -m -u 1000 user
-
-# Switch to the "user" user
-USER user
-
-# Set home to the user's home directory
-ENV HOME=/home/user \
-	PATH=/home/user/.local/bin:$PATH
-
-
 WORKDIR $HOME/app
 
-
-#RUN chmod 777 /data
-#RUN useradd -m -u 1000 user
-#USER user
 # Copy the current directory contents into the container at $HOME/app setting the owner to the user
 COPY --chown=user . $HOME/app
-
-ENV AMBERHOME="/usr/bin/amber22"
-ENV PATH="$AMBERHOME/bin:$PATH"
-ENV PYTHONPATH="$AMBERHOME/lib/python3.8/site-packages"
-
-RUN pip install -r requirements.txt
 CMD ["python", "main.py"]
 
 

README.md

@@ -1,5 +1,5 @@
 ---
-title: Adaptability protein dynamics
+title: Adapt Docker
 emoji: 🔥
 colorFrom: indigo
 colorTo: red

main.py

@@ -1,7 +1,7 @@
-
-
 import gradio as gr
+
 import py3Dmol
+
 from Bio.PDB import *
 
 import numpy as np
@@ -12,9 +12,6 @@ import os
 from MDmodel import GNN_MD
 import h5py
 from transformMD import GNNTransformMD
-import sys
-import pytraj as pt
-import pickle 
 
 # JavaScript functions
 resid_hover = """function(atom,viewer) {{
@@ -49,78 +46,6 @@ model = model.to('cpu')
 model.eval()
 
 
-def run_leap(fileName, path):
-    leapText = """
-    source leaprc.protein.ff14SB
-    source leaprc.water.tip3p
-    exp = loadpdb PATH4amb.pdb
-    saveamberparm exp PATHexp.top PATHexp.crd
-    quit
-    """
-    with open(path+"leap.in", "w") as outLeap:
-        outLeap.write(leapText.replace('PATH', path))	
-    os.system("tleap -f "+path+"leap.in >> "+path+"leap.out")
-
-def convert_to_amber_format(pdbName):
-    fileName, path = pdbName+'.pdb', ''
-    os.system("pdb4amber -i "+fileName+" -p -y -o "+path+"4amb.pdb -l "+path+"pdb4amber_protein.log")
-    run_leap(fileName, path)
-    traj = pt.iterload(path+'exp.crd', top = path+'exp.top')
-    pt.write_traj(path+fileName, traj, overwrite= True)
-    print(path+fileName+' was created. Please always use this file for inspection because the coordinates might get translated during amber file generation and thus might vary from the input pdb file.')
-    return pt.iterload(path+'exp.crd', top = path+'exp.top')
-
-def get_maps(mapPath):
-    residueMap = pickle.load(open(os.path.join(mapPath,'atoms_residue_map_generate.pickle'),'rb'))
-    nameMap = pickle.load(open(os.path.join(mapPath,'atoms_name_map_generate.pickle'),'rb'))
-    typeMap = pickle.load(open(os.path.join(mapPath,'atoms_type_map_generate.pickle'),'rb'))
-    elementMap = pickle.load(open(os.path.join(mapPath,'map_atomType_element_numbers.pickle'),'rb'))
-    return residueMap, nameMap, typeMap, elementMap
-
-def get_residues_atomwise(residues):
-    atomwise = []
-    for name, nAtoms in residues:
-        for i in range(nAtoms):
-            atomwise.append(name)
-    return atomwise
-
-def get_begin_atom_index(traj):
-    natoms = [m.n_atoms for m in traj.top.mols]
-    molecule_begin_atom_index = [0] 
-    x = 0
-    for i in range(len(natoms)):
-        x += natoms[i]
-        molecule_begin_atom_index.append(x)
-    print('molecule begin atom index', molecule_begin_atom_index, natoms)
-    return molecule_begin_atom_index
-
-def get_traj_info(traj, mapPath):
-    coordinates  = traj.xyz
-    residueMap, nameMap, typeMap, elementMap = get_maps(mapPath)
-    types = [typeMap[a.type] for a in traj.top.atoms]
-    elements = [elementMap[typ] for typ in types]
-    atomic_numbers = [a.atomic_number for a in traj.top.atoms]
-    molecule_begin_atom_index = get_begin_atom_index(traj)
-    residues = [(residueMap[res.name], res.n_atoms) for res in traj.top.residues]
-    residues_atomwise = get_residues_atomwise(residues)
-    return coordinates[0], elements, types, atomic_numbers, residues_atomwise, molecule_begin_atom_index
-
-def write_h5_info(outName, struct, atoms_type, atoms_number, atoms_residue, atoms_element, molecules_begin_atom_index, atoms_coordinates_ref):
-    if os.path.isfile(outName):
-        os.remove(outName)
-    with h5py.File(outName, 'w') as oF:
-        subgroup = oF.create_group(struct)     
-        subgroup.create_dataset('atoms_residue', data= atoms_residue, compression = "gzip", dtype='i8')
-        subgroup.create_dataset('molecules_begin_atom_index', data= molecules_begin_atom_index, compression = "gzip", dtype='i8')
-        subgroup.create_dataset('atoms_type', data= atoms_type, compression = "gzip", dtype='i8')
-        subgroup.create_dataset('atoms_number', data= atoms_number, compression = "gzip", dtype='i8')  
-        subgroup.create_dataset('atoms_element', data= atoms_element, compression = "gzip", dtype='i8')
-        subgroup.create_dataset('atoms_coordinates_ref', data= atoms_coordinates_ref, compression = "gzip", dtype='f8')
-
-def preprocess(pdbid: str = None, ouputfile: str = "inference_for_md.hdf5", mask: str = "!@H=", mappath: str = "/maps/"):
-    traj = convert_to_amber_format(pdbid)
-    atoms_coordinates_ref, atoms_element, atoms_type, atoms_number, atoms_residue, molecules_begin_atom_index = get_traj_info(traj[mask], mappath)
-    write_h5_info(ouputfile, pdbid, atoms_type, atoms_number, atoms_residue, atoms_element, molecules_begin_atom_index, atoms_coordinates_ref)
 
 def get_pdb(pdb_code="", filepath=""):
     try:
@@ -140,37 +65,24 @@ def get_offset(pdb):
             return int(line[22:27])
 
 
-def get_pdbid_from_filename(filename: str):
-    # Assuming the filename would be of the standard form 11GS.pdb
-    return filename.split(".")[0]
-
-def predict(pdb_code, pdb_file, topN):
-    #path_to_pdb = get_pdb(pdb_code=pdb_code, filepath=pdb_file)
-
-    #pdb = open(path_to_pdb, "r").read()
-    # switch to misato env if not running from container
-
-    pdbid = get_pdbid_from_filename(pdb_file.name)
+def predict(pdb_code, pdb_file):
+    path_to_pdb = get_pdb(pdb_code=pdb_code, filepath=pdb_file)
     mdh5_file = "inference_for_md.hdf5"
-    mappath = "/maps"
-    mask = "!@H="
-    preprocess(pdbid=pdbid, ouputfile=mdh5_file, mask=mask, mappath=mappath)
-
     md_H5File = h5py.File(mdh5_file)
 
     column_names = ["x", "y", "z", "element"]
     atoms_protein = pd.DataFrame(columns = column_names)
-    cutoff = md_H5File[pdbid]["molecules_begin_atom_index"][:][-1] # cutoff defines protein atoms
+    cutoff = md_H5File["11GS"]["molecules_begin_atom_index"][:][-1] # cutoff defines protein atoms
 
-    atoms_protein["x"] = md_H5File[pdbid]["atoms_coordinates_ref"][:][:cutoff, 0]
-    atoms_protein["y"] = md_H5File[pdbid]["atoms_coordinates_ref"][:][:cutoff, 1]
-    atoms_protein["z"] = md_H5File[pdbid]["atoms_coordinates_ref"][:][:cutoff, 2]
+    atoms_protein["x"] = md_H5File["11GS"]["atoms_coordinates_ref"][:][:cutoff, 0]
+    atoms_protein["y"] = md_H5File["11GS"]["atoms_coordinates_ref"][:][:cutoff, 1]
+    atoms_protein["z"] = md_H5File["11GS"]["atoms_coordinates_ref"][:][:cutoff, 2]
 
-    atoms_protein["element"] = md_H5File[pdbid]["atoms_element"][:][:cutoff]  
+    atoms_protein["element"] = md_H5File["11GS"]["atoms_element"][:][:cutoff]  
 
     item = {}
     item["scores"] = 0
-    item["id"] = pdbid
+    item["id"] = "11GS"
     item["atoms_protein"] = atoms_protein
 
     transform = GNNTransformMD()
@@ -184,56 +96,30 @@ def predict(pdb_code, pdb_file, topN):
     for i in range(adaptability.shape[0]):
         data.append([i, atom_mapping[atoms_protein.iloc[i, atoms_protein.columns.get_loc("element")] - 1], atoms_protein.iloc[i, atoms_protein.columns.get_loc("x")],atoms_protein.iloc[i, atoms_protein.columns.get_loc("y")],atoms_protein.iloc[i, atoms_protein.columns.get_loc("z")],adaptability[i]])
 
+    topN = 100
     topN_ind = np.argsort(adaptability)[::-1][:topN]    
 
-    pdb = open(pdb_file.name, "r").read()
-    pdb2 = pdb
-    
-    view = py3Dmol.view(width=1000, height=800)
+    pdb = open(path_to_pdb, "r").read()
+
+    view = py3Dmol.view(width=600, height=400)
     view.setBackgroundColor('white')
     view.addModel(pdb, "pdb")
-    view.setStyle({'stick': {'colorscheme': {'prop': 'resi', 'C': '#cccccc'}},'cartoon': {'color': '#4c4e9e', 'alpha':"0.6"}})
-
-    #view.addModel(pdb2, "pdb2")
-    #view.setStyle({'cartoon': {'color': 'gray'}})
-
-
-    # Commenting since the visualizer is not rendered
-    # view.addLight([0, 0, 10], [1, 1, 1], 1)  # Add directional light from the z-axis
-    # view.setSpecular(0.5)  # Adjust the specular lighting effect
-    # view.setAmbient(0.5)  # Adjust the ambient lighting effect
-    
+    view.setStyle({'stick': {'colorscheme': {'prop': 'resi', 'C': 'turquoise'}}})
+   
     for i in range(topN):
-        adaptability_value = adaptability[topN_ind[i]]
-        color = '#a0210f'
-        view.addSphere({
-            'center': {
-                'x': atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("x")],
-                'y': atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("y")],
-                'z': atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("z")]
-            },
-            'radius': adaptability_value / 1.5,
-            'color': color,
-            'alpha': 0.75
-        })
-
-           
+        view.addSphere({'center':{'x':atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("x")], 'y':atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("y")],'z':atoms_protein.iloc[topN_ind[i], atoms_protein.columns.get_loc("z")]},'radius':adaptability[topN_ind[i]]/1.5,'color':'orange','alpha':0.75})    
+
     view.zoomTo()
 
     output = view._make_html().replace("'", '"')
 
     x = f"""<!DOCTYPE html><html> {output} </html>"""  # do not use ' in this input
-    
-    return f"""<iframe  style="width: 100%; height:820px" name="result" allow="midi; geolocation; microphone; camera; 
+    return f"""<iframe  style="width: 100%; height:420px" name="result" allow="midi; geolocation; microphone; camera; 
     display-capture; encrypted-media;" sandbox="allow-modals allow-forms 
     allow-scripts allow-same-origin allow-popups 
     allow-top-navigation-by-user-activation allow-downloads" allowfullscreen="" 
     allowpaymentrequest="" frameborder="0" srcdoc='{x}'></iframe>""", pd.DataFrame(data, columns=['index','element','x','y','z','Adaptability'])
 
-def export_csv(d):
-    d.to_csv("adaptabilities.csv")
-    return gr.File.update(value="adaptabilities.csv", visible=True)
-
 
 callback = gr.CSVLogger()
 
@@ -244,11 +130,7 @@ def run():
         #text_input = gr.Textbox()
         #text_output = gr.Textbox()
         #text_button = gr.Button("Flip")
-        inp = gr.Textbox(placeholder="Upload PDB file below", label="Input structure")
-        #inp = ""
-        topN = gr.Slider(value=100,
-            minimum=1, maximum=1000, label="Number of highest adaptability values to visualize", step=1
-        )
+        inp = gr.Textbox(placeholder="PDB Code or upload file below", label="Input structure")
         pdb_file = gr.File(label="PDB File Upload")
         #with gr.Row():
         #    helix = gr.ColorPicker(label="helix")
@@ -257,21 +139,14 @@ def run():
         single_btn = gr.Button(label="Run")
         with gr.Row():
             html = gr.HTML()
-        with gr.Row():
-            Dbutton = gr.Button("Download adaptability values")
-            csv = gr.File(interactive=False, visible=False)
         with gr.Row():
             dataframe = gr.Dataframe()
                 
-        single_btn.click(fn=predict, inputs=[inp, pdb_file, topN], outputs=[html, dataframe])
-
-        Dbutton.click(export_csv, dataframe, csv)
-
-        
+        single_btn.click(fn=predict, inputs=[inp, pdb_file], outputs=[html, dataframe])
 
 
     demo.launch(server_name="0.0.0.0", server_port=7860)
 
 
 if __name__ == "__main__":
-    run()
+    run()

maps/atoms_name_map_for_pdb.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:43312dc497c280abd0aa8aa5ca5e70cb33aef74348d880eabef62e919c8fa04c
-size 26765

maps/atoms_name_map_generate.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:16d45a0c1413b2ef3f1e5f044dd13234d5f1fadb5cf5f09b4af528872b83808b
-size 4264

maps/atoms_name_map_new.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:becf7b36cb63e74072d0218d19ee8c1187e19e6d7f7189409e95b00e6bf5c4a8
-size 4264

maps/atoms_residue_map.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e8cc11f197830509123ba01140796a8a032ca0630e0e93962391dcdeed38a40a
-size 284

maps/atoms_residue_map_generate.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a10a0235a59670f8891c5d4d6dc47c474625fc3d5fbeb7161d3a34341ba23f7a
-size 284

maps/atoms_type_map.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3dedc6173b3889c25bb05dacc5ea90d057b6ac7758181eb9465f85c79aafca5d
-size 1207

maps/atoms_type_map_generate.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:98a595ca5369902304c7c10a3d2c4d97c754d40f70d88b6aa9f725f186587c96
-size 1207

maps/map_atomType_element_names.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5c1900a90db246fee15e600a30c5ae40879f04f67d7d42326dd8680df7f74383
-size 356

maps/map_atomType_element_numbers.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:bb6cf66a0f83ff33af562a8c1d6f716abb857b99a3137313fe86e24936afe39c
-size 448

maps/map_elements_numbers_number.pickle

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:67cc394f5216b325880af8cdb3919b0634fe5a5bf1598bbfb1985a66d111f7f6
-size 110

mock_download.py

@@ -1,23 +0,0 @@
-import gradio as gr
-
-def mock_ocr(f):
-    return [[1, 2, 3], [4, 5, 6]]
-
-def export_csv(d):
-    d.to_csv("output.csv")
-    return gr.File.update(value="output.csv", visible=True)
-
-with gr.Blocks() as demo:
-    with gr.Row():
-        file = gr.File(label="PDF file", file_types=[".pdf"])
-        dataframe = gr.Dataframe()
-        
-        with gr.Column():
-            button = gr.Button("Export")
-            csv = gr.File(interactive=False, visible=False)
-            
-            
-    file.change(mock_ocr, file, dataframe)
-    button.click(export_csv, dataframe, csv)
-        
-demo.launch()

old_main.py

@@ -0,0 +1,30 @@
+import gradio as gr
+import torch
+import requests
+from torchvision import transforms
+
+model = torch.hub.load("pytorch/vision:v0.6.0", "resnet18", pretrained=True).eval()
+response = requests.get("https://git.io/JJkYN")
+labels = response.text.split("\n")
+
+
+def predict(inp):
+    inp = transforms.ToTensor()(inp).unsqueeze(0)
+    with torch.no_grad():
+        prediction = torch.nn.functional.softmax(model(inp)[0], dim=0)
+        confidences = {labels[i]: float(prediction[i]) for i in range(1000)}
+    return confidences
+
+
+def run():
+    demo = gr.Interface(
+        fn=predict,
+        inputs=gr.inputs.Image(type="pil"),
+        outputs=gr.outputs.Label(num_top_classes=3),
+    )
+
+    demo.launch(server_name="0.0.0.0", server_port=7860)
+
+
+if __name__ == "__main__":
+    run()

requirements.txt

@@ -1,6 +1,3 @@
 gradio
 torchvision
-requests
-py3Dmol
-biopython
-pandas
+requests